Struct ReplayExecutionConfig 

pub struct ReplayExecutionConfig {
    pub skip_puts: bool,
    pub skip_finds: bool,
    pub skip_asks: bool,
    pub stop_on_mismatch: bool,
    pub verbose: bool,
    pub top_k: Option<usize>,
    pub adaptive: bool,
    pub min_relevancy: f32,
    pub audit_mode: bool,
    pub use_model: Option<String>,
    pub generate_diff: bool,
}

Configuration for replay execution.

Fields

skip_puts: bool

Skip put actions (useful for read-only replay)

skip_finds: bool

Skip find actions

skip_asks: bool

Skip ask actions (useful when LLM not available)

stop_on_mismatch: bool

Stop on first mismatch

verbose: bool

Verbose logging

top_k: Option<usize>

Override top-k for find actions (None = use original values) Higher values reveal documents that may have been missed

adaptive: bool

Use adaptive retrieval based on score distribution

min_relevancy: f32

Minimum relevancy score for adaptive mode (0.0-1.0)

audit_mode: bool

AUDIT MODE: Use frozen retrieval for ASK actions When true, ASK actions use recorded frame IDs instead of re-executing search

use_model: Option<String>

Override model for audit replay (format: “provider:model”) When set, re-executes LLM call with frozen context using this model

generate_diff: bool

Generate diff report comparing original vs new answers

Trait Implementations

impl Clone for ReplayExecutionConfig

fn clone(&self) -> ReplayExecutionConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ReplayExecutionConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for ReplayExecutionConfig

fn default() -> Self

Returns the “default value” for a type.